An RNA Helicase DHX33 Inhibitor Shows Broad Anticancer Activity via Inducing Ferroptosis in Cancer Cells

Abstract

RNA helicase DHX33 has been identified as a critical factor promoting cancer development. In the present study, a previously developed small molecule inhibitor for DHX33, KY386, was found to robustly kill cancer cells via a new path, the ferroptosis pathway. Mechanistically, DHX33 promotes the expression of critical players in lipid metabolism including FADS1, FADS2, and SCD1 genes, thereby sensitizing cancer cells to ferroptosis mediated cell death. Our study reveals a novel mechanism of DHX33 in promoting tumorigenesis and highlights that pharmacological targeting DHX33 can be a feasible option in human cancers. Normally differentiated cells are insensitive to DHX33 inhibition, and DHX33 inhibitors have little cellular toxicity in vitro and in vivo. Our studies demonstrated that DHX33 inhibitors can be promising anticancer agents with great potential for cancer treatment.

Figure 1

KY386 shows broad anticancer activity. A. 38 different cancer cell lines were treated with KY386 at multiple doses for 72 h, and ATP content was measured and used as a surrogate of cell viability using the CellTiter-Glo assay. Three independent experiments were repeated, and each individual IC₅₀ was listed in the table. B. CCK-8 analysis was performed on the designated cancer cell lines for KY386 efficacy in vitro. Cells were treated at multiple doses for 72 h. Taxol was used as a positive control for comparison. The calculated IC₅₀ was labeled in the graph. C. Three human cancer organoids, each of metastatic pancreatic cancer and lung cancer with RAS gene mutation (G12 V or G12D) were used for the KY386 drug potency test. Cell viability was plotted against the concentration of KY386 (in log values). IC₅₀ was calculated and shown.

Figure 2

Normally differentiated cells and cancer cells with low DHX33 levels are less sensitive to KY386 inhibition. A. Different cancer cells with the same tissue origin were subjected to western blot analysis with the DHX33 antibody; GAPDH was used as an internal control. For controls, several human normal cells were also analyzed. B. CCK-8 analysis was performed on a few human normal cell lines for KY386 efficacy in vitro. Cells were treated at multiple doses for 72 h. The calculated IC₅₀ was labeled in the graph.

Figure 3

KY386 induced apoptosis in some cancer cells but much less efficient in others. A. Cell apoptosis analysis was performed in two representative cancer cell lines, HuH-7 and HCT116, and doxorubicin served as a positive control. KY386 was used at different concentrations. Cell apoptosis was analyzed based on Annexin V/PI staining from the standard protocol of Cell Vybrant kit. The apoptosis index was plotted. B. The dot plot for apoptosis analysis in (A). C.HGC27 and DU145 cells were treated with KY386 at different doses for the designated time points. Cell apoptosis analysis was then performed following the standard protocol of a Cell Vybrant kit. The apoptosis index was evaluated based on Annexin V staining only and was labeled on the graph.

Figure 4

KY386 induced ferroptosis in a broad spectrum of cancer cells. A. HGC27 cells were treated with KY386 or inducer of ferroptosis (provided in the kit) at the indicated doses for either 8 or 16 h. Cells were then analyzed for ROS or GSH levels according to the standard protocol provided by the manufacture. Data reflected mean and standard deviation from three separate experiments. *, p < 0.05. B. Similarly, ROS levels were analyzed in other cancer cell lines after KY386 treatment for either 4, 6, 8 or 24 h. *, p < 0.05.C. Lipid peroxidation (LPO) levels were evaluated based on the standard protocol recommended by the manufacturer after cancer cells were treated with KY386 at different doses for the indicated time. *, p < 0.05. D. Fe²⁺ levels were evaluated based on the standard protocol recommended by the manufacturer after cancer cells were treated with KY386 at different doses for the indicated time. *, p < 0.05.

Figure 5

DHX33 promotes the expression of critical players in lipid metabolism including FADS1, FADS2 and SCD1. A. Cancer cells were treated with KY386 at the indicated doses for 24 h. Whole cell extracts were prepared and were subjected to western blot analysis with the indicated antibodies. GAPDH served as an internal control. B. Cancer cells were transduced by lentivirus to knockdown DHX33, and shScrambled (shSCR) was used as a control. Whole cell extracts were then analyzed by western blot with the indicated antibodies, and GAPDH serves as an internal control. C. Human normal cells HSF or BJ cells were either treated with KY386 at the indicated doses for 24 h, followed by whole cell lysates preparation, or were infected by lentivirus to knockdown DHX33, shScrambled was used as a control. Three days post infection, whole cell extracts were prepared. Western blot analysis was performed on these cell lysates with designated antibodies. GAPDH served as a control.

Figure 6

DHX33 regulates fads1, fads2, and scd1 genes at the transcription level. A. A875, A375, and HGC27 cells were treated with KY386 at 40 nM for the indicated time points. Cells were then harvested, and total RNA was extracted for QPCR analysis with the indicated primer sets. Experiments were repeated at least twice, *, p < 0.05. B. T24 and DU145 cells were first treated with KY386 at the indicated doses for 6 h and then harvested for QPCR analysis. These cells were then analyzed for target gene expression after treatment with KY386 at 40 nM for the indicated time points. Total RNA was extracted for QPCR analysis. Experiments were repeated at least twice, *, p < 0.05. C. HGC27 cells were infected with lentivirus encoding either shSCR or shDHX33 to knockdown DHX33. 72 h postinfections, cells were harvested for total RNA extraction and QPCR analysis. Experiments were repeated at least twice, *, p < 0.05.

Figure 7

KY386 inhibits the growth of various human cancers in vivo. A. The SNU668 gastric xenograft model was set up, and mice were subjected to drug treatment through intraperitoneal injection. Tumor volume was monitored and calculated. B. Tumor images were taken after tumor dissection. C. Mouse body weight was monitored and plotted. D. Patient derived xenograft models for Ras G12D mutant colon cancers were set up and mice were subjected to drug treatment through intraperitoneal injection. Tumor volume was monitored and calculated. E. Tumor images were taken after tumor dissection. F. Mouse body weight was monitored and plotted.